Functional and structural changes in the visual pathway in multiple sclerosis.

INTRODUCTION
Multiple sclerosis (MS) is a heterogeneous disease with an unpredictable course. Visual pathway is a target of the disease and may reflect mechanisms that lead to disability. Structural and functional changes in the visual pathway may be studied by noninvasive techniques such as optical coherence tomography (OCT), visual evoked potentials (VEP), or B-mode transorbital sonography (TOS).


OBJECTIVES
The aim is to assess changes in the visual pathway in eyes of MS patients with and without a history of optic neuritis over a 3-year period and to explore their relationship with disability.


MATERIALS AND METHODS
In total, 112 eyes from 56 patients with relapsing MS were recruited: 29 with, and 83 without a history of ON (hON and nhON, respectively). Several parameters were measured by OCT, VEP, and TOS. Baseline measurements were also compared to 29 healthy controls. At 36 months, measurements were repeated in all eyes.


RESULTS
At baseline, all tests showed significant differences in optic nerve structure and function in both patient cohorts in all the parameters studied, suggestive of more impairment of the visual pathway among the hON cohort. OCT showed significant differences between healthy controls and the nhON cohort. At 36 months, the nhON cohort showed significant changes by OCT, VEP, and TOS suggestive of further visual pathway impairment. OCT measurements also correlated with baseline EDSS among the nhON cohort.


CONCLUSIONS
OCT is the most suitable technique and outperforms VEP and TOS to detect subclinical damage in the visual pathway. It discriminated MS patients from healthy controls and showed a progressive decline in optic nerve thickness over time among these patients.


| INTRODUC TI ON
Multiple sclerosis (MS) is a heterogeneous disease with a variable clinical course. Much research has been focused on biomarkers that could predict patients' clinical course and long-term disability.
Study of the visual pathway has been a field of interest as it may reflect brain changes that contribute to patient disability (Costello, 2013).
The retina and the optic nerve are functionally eloquent structures, known to be important targets of the disease even in patients without optic neuritis (ON). Their study may portray brain axonal damage, with the advantage of being easily accessible by inexpensive and noninvasive techniques such as optical coherence tomography (OCT), visual evoked potentials (VEP), and transorbital sonography (TOS) of the retrobulbar optic nerve. The aim of the study was to evaluate the suitability of these three techniques to monitor visual pathway changes over a 3-year period in MS patients and to explore their relationship with disability.

| PATIENTS AND ME THODS
This was a prospective, 36-month observational study of a total of 112 eyes from 56 patients with MS, of which 29 had history and 27 did not have a history of ON. Patients were recruited from the MS clinic in the Neurology Service at Hospital General Universitario de Valencia. Eyes of patients were segregated into two cohorts according to the history of optic neuritis (hON eye cohort, n = 29) or non-history of ON (nhON eye cohort, n = 83).
All eyes underwent VEP, OCT, and TOS measurements at baseline and at 36 months. Correlation between these findings and clinical variables was explored. At the beginning of the study, baseline measurements were also compared to a healthy group.
To ensure comparability of the cohorts, baseline demographic and clinical variables of the patients were collected and compared before eye segregation. These included disease duration, accumulated number of relapses, baseline neurological disability as measured by the EDSS, baseline MRI T2 lesion and T1 Gd-enhancing lesion count, and the presence of disease-modifying treatment (Table 1).

| Patients
We included patients younger than 50 years with relapsing-remitting MS (RRMS) diagnosis according to the 2005 McDonald criteria (Polman et al., 2005) with a score on the Expanded Disability Status Scale (EDSS) of less than or equal to 4. Measurements were taken at least 6 months after last ON episode. Patients with causes of vision loss or retinal damage not attributable to MS were excluded ( Figure 1).
Among the hON cohort, recordings were taken at least 6 months after the onset of ON symptoms. ON was diagnosed by the clinical criteria of ocular pain with eye movements and a decrease in visual acuity referred by the patient and confirmed in the ophthalmological examination. During the acute phase of ON, all the patients received 1 g of intravenous methylprednisolone for 3-5 days.
The study was completed ensuring compliance with current ethical and legal standard (Declaration of Helsinki) and approved by the local ethics and research committee. All the participants signed their informed consent prior to inclusion in the study.

| Instrumental variables measured
A set of nine measurements were taken for each eye. These were latency (VEP-lat) and amplitude (VEP-amp) of the p100 wave of the VEP, global and quadrant-based (inferior, nasal, superior, temporal) retinal nerve fiber layer thickness (RNFL-thick, I-RNFL-thick, N-RNFL-thick, S-RNFL-thick, and T-RNFL-thick, respectively) measured by OCT, longitudinal diameter (LD), and transversal diameter (TD) of the retrobulbar optic nerve measured by TOS.

| Transorbital ultrasound
Transorbital ultrasound (TOS) measurements of the retrobulbar optic nerve were taken with a color duplex ultrasound with a 10 Hz linear probe (Esaote MyLab™ 25 Gold). To measure the LD, the transducer was placed on the upper eyelid and the measurement was taken 3 mm from the papilla; to measure the TD, the probe was placed on the lower eyelid, perpendicular to the facial plane.

| OCT
RNFL-thick was measured by a spectral domain OCT (SD 3D OCT 2000 Top Con) with a 3D acquisition protocol (6.0 × 6.0 mm-512 × 128). Patients were scanned under photopic conditions with ceiling lights and without pupil dilation. An experienced grader masked for clinical presentation checked all images for sufficient image quality in line with the OSCAR-IB criteria (Schippling et al., 2015) and manually corrected segmentation in case of errors. All scans were carefully checked by an ophthalmologist.

| Statistics
Patients who met the inclusion criteria were consecutively recruited.

| Baseline
The baseline clinical and demographic characteristics of the three groups are summarized in

| Diagnostic performance of OCT, VEP, and TOS
To evaluate the ability of OCT, VEP, and TOS to discriminate between nhON from controls (and thus, the capability of detecting subclinical damage), we carried out a receiver operating characteristic (ROC) analysis for each test (Figure 1)    Alteration of VEPs in patients without a history of ON may be useful as a supporting criterion in the diagnosis of MS because these changes may indicate subclinical damage in the visual pathway. Additionally, the functional deterioration we observed at the 3-year follow-up suggests that there is progressive functional damage to this pathway, even when visually asymptomatic. To corroborate these data and obtain more robust results, further studies with extended follow-up periods and/or a larger sample size would be desirable. Also, the use of multifocal VEPs with higher sensitivity and specificity would be very useful in detecting these visual pathway abnormalities (Klistorner et al., 2008;Laron et al., 2009).

| Comparison of baseline and 36-month followup measurements
Given the data above, we conclude that OCT, VEP, and TOS may be valid to monitor visual pathway damage in MS patients and that OCT is the most sensitive and specific when compared to the other two. Considering the hypothesis that axonal loss is the predominant cause of disability in MS, it is predicted that RNFL thickness should be smaller in the progressive subtypes of the disease, which has already been shown by several authors (Alonso et al., 2018;Balk et al., 2016;Henderson et al., 2008;Pulicken et al., 2007;Ratchford et al., 2013). Interestingly, the rate of RNFL thinning has been found to be greater in active and earlier phases of the disease, which could reflect a plateau effect of advanced stages (Klistorner et al., 2008;Laron et al., 2009). Future studies intending to use any of these techniques as a measure of disability should be aware of eyes previously affected by an episode of ON, since baseline measurements could be already affected and thus, not reflect the overall axonal damage present. In our series, despite progressive damage to the optic nerve, no correlations were found between final EDSS and any of the parameters studied during follow-up. Therefore, we cannot conclude that increased axonal damage corresponds to a higher disability. Similarly, other authors were also unable to demonstrate this association (Henderson et al., 2008;Oreja-Guevara, Noval, Manzano, & Diez-Tejedor, 2010), although recent studies established a relationship between RNFL thickness and risk of disability progression (Martínez-Lapiscina et al., 2016), and rate of RNFL thinning and NEDA-3 at 2 years (AUC = 0.8) (Pisa et al., 2017) There may be multiple reasons for this, including the short follow-up period, patient heterogeneity, the small sample sizes used in most studies, or perhaps because EDSS is not sensitive enough to detect changes in disability when there is no motor impairment. Therefore, future studies should aim to use larger sample sizes, longer follow-up times, and more sensitive scales capable of detecting subtler changes in disability.

ACK N OWLED G M ENTS
Novartis for financial support.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to disclose.

AUTH O R CO NTR I B UTI O N
Carcelén-Gadea M made substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data and was involved in drafting the manuscript or revising it critically for important intellectual content. Quintanilla-Bordás C, Gracia-García A, García-Villanueva C, Jannone-Pedro N, Alvarez-Sánchez L, Vilaplana-Domínguez L, Blanco-Hernández T, Pons-Amate JM, and Cervelló-Donderis A contributed to acquisition of data and was involved in drafting the manuscript or revising it critically for important intellectual content.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.